[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124624-en":3,"doc-seo-124624-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},124624,687197100911,"Himbo","https://ap-avatar.wpscdn.com/avatar/a000239b6f1da00475?x-image-process=image/resize,m_fixed,w_180,h_180&k=1785132997149421697",8,"Research & Report","CRUNCHGPT - A CHATGPT ASSISTED FRAMEWORK FOR SCIENTIFIC MACHINE LEARNING","Scientific Machine Learning (SciML) integrates data and physics to accelerate computational science and engineering, yet key steps—preprocessing, problem formulation, code generation, postprocessing, and analysis—remain time-consuming and limit adoption in industrial settings and digital twin workflows. CrunchGPT unifies these stages under ChatGPT, acting as a workflow conductor driven by simple user prompts. Two interactive examples cover aerodynamic airfoil optimization and flow-field computation using validation-focused workflows, supported by a guided web interface. The framework is designed to extend toward broader computational mechanics and educational and research use.","CRUNCHGPT: A CHATGPT ASSISTED FRAMEWORK FOR  \nSCIENTIFIC MACHINE LEARNING  \narXiv :2306 . 15551v1 [ cs .LG] 27 Jun 2023  \nVarun Kumar 􀀃 [varun_kumar2@brown.edu](varun_kumar2@brown.edu)  \nLeonard Gleyzer y [leonard_gleyzer@brown.edu](leonard_gleyzer@brown.edu)  \nAdar Kahana y[adar_kahana@brown.edu](adar_kahana@brown.edu)  \nKhemraj Shukla y [khemraj_shukla@brown.edu](khemraj_shukla@brown.edu)  \nGeorge Em Karniadakis 􀀃 y z  \n[george_karniadakis@brown.edu](george_karniadakis@brown.edu)  \nABSTRACT  \nScientific Machine Learning (SciML) has advanced recently across many different areas in computational science and engineering. The objective is to integrate data and physics seamlessly without the need of employing elaborate and computationally taxing data assimilation schemes. However, preprocessing, problem formulation, code generation, postprocessing and analysis are still timeconsuming and may prevent SciML from wide applicability in industrial applications and in digital twin frameworks. Here, we integrate the various stages of SciML under the umbrella of ChatGPT, to formulate CrunchGPT, which plays the role of a conductor orchestrating the entire workflow of SciML based on simple prompts by the user. Specifically, we present two examples that demonstrate the potential use of CrunchGPT in optimizing airfoils in aerodynamics, and in obtaining flow fields in various geometries in interactive mode, with emphasis on the validation stage. To demonstrate the flow of the CrunchGPT, and create an infrastructure that can facilitate a broader vision, we built awebapp based guided user interface, that includes options for a comprehensive summary report. The overall objective is to extend CrunchGPT to handle diverse problems in computational mechanics, design, optimization and controls, and general scientific computing tasks involved in SciML, hence using it as a research assistant tool but also as an educational tool. While here the examples focus in fluid mechanics, future versions will target solid mechanics and materials science, geophysics, systems biology and bioinformatics.  \nKeywords ChatGPT for design 􀀁 NACA airfoil optimization 􀀁 ChatGPT for PINNs 􀀁 fluid mechanics 􀀁 cavity flow 􀀁 validation  \n1 Introduction  \nWith the advancements in deep learning and natural language processing techniques in recent times, Large Language Models (LLMs) have witnessed remarkable development and research interest, both from academia and industry alike. The availability of large text corpus in the form of digital content, coupled with modern language processing techniques such as Transformers [1], has improved the efficacy and accuracy of natural language processing tasks such as language translation, text classification and sentence completion. Some state-of-the-art LLMs include BERT [2], T5 [3], RoBERTa [4], XLNet [5] amongst others. These models are typically trained on large text corpus, contain billions of neural network parameters, and can be adapted to meet a user's specific use case using these pre-trained models [6–8] . Asignificant and potentially breakthrough advancement was achieved by OpenAI through the development of Generative Pre-trained Transformer (GPT-3), which has shown remarkable performance for a range of natural language processing tasks ranging from sentence completion to providing specific responses based on instructions provided by the user [9] . Colloquially named as ChatGPT, the GPT language model is one of the largest natural language processing models  \n􀀃 School of Engineering, Brown University, Providence, RI.  \nyDivision of Applied Mathematics, Brown University, Providence, RI. zCorresponding author  \nCrunchGPT: SciML Assistant  \n(with hundreds of billions of parameters) that exists today and is capable of generating very high-quality, human-like responses to user queries on a wide range of topics.  \nA key ability of ChatGPT model that makes it a significant improvement over existing language models is its ability to learn and comprehen","cbCaiiHBWDAdJ6oq","https://ap.wps.com/l/cbCaiiHBWDAdJ6oq","pdf",8067535,1,23,"English","en",105,"# Introduction\n## Large Language Models and ChatGPT\n## Motivation for SciML workflow automation\n## CrunchGPT overview and two demonstrated tasks","[{\"question\":\"What problem does CrunchGPT address in SciML workflows?\",\"answer\":\"CrunchGPT targets the time-consuming steps of preprocessing, problem formulation, code generation, postprocessing, and analysis that currently hinder wide SciML adoption in industrial applications and digital twins.\"},{\"question\":\"How does CrunchGPT work from the user’s perspective?\",\"answer\":\"Users provide simple prompts, and CrunchGPT orchestrates the end-to-end SciML workflow. It also supports interactive modes for running and validating tasks.\"},{\"question\":\"What examples demonstrate CrunchGPT’s potential?\",\"answer\":\"The document presents two examples: optimizing 2D NACA airfoils in aerodynamics and obtaining flow fields for various geometries using PINNs, with emphasis on validation. It also describes a webapp interface with options for comprehensive summary reporting.\"}]","CRUNCHGPT - A CHATGPT ASSISTED FRAMEWORK FOR SCIENTIFIC MACHINE LEARNING | PDF",1785893379,58,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"crunchgpt-a-chatgpt-assisted-framework-for-scientific-machine-learning","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/crunchgpt-a-chatgpt-assisted-framework-for-scientific-machine-learning/124624/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does CrunchGPT address in SciML workflows?","Question",{"text":75,"@type":76},"CrunchGPT targets the time-consuming steps of preprocessing, problem formulation, code generation, postprocessing, and analysis that currently hinder wide SciML adoption in industrial applications and digital twins.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does CrunchGPT work from the user’s perspective?",{"text":80,"@type":76},"Users provide simple prompts, and CrunchGPT orchestrates the end-to-end SciML workflow. It also supports interactive modes for running and validating tasks.",{"name":82,"@type":73,"acceptedAnswer":83},"What examples demonstrate CrunchGPT’s potential?",{"text":84,"@type":76},"The document presents two examples: optimizing 2D NACA airfoils in aerodynamics and obtaining flow fields for various geometries using PINNs, with emphasis on validation. It also describes a webapp interface with options for comprehensive summary reporting.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]